[whatmodel.sh](../../README.md) / DeepSeek-Coder 33B Base
# DeepSeek-Coder 33B Base
DeepSeek · released 2023-10 · 33.34B params · deepseek
**score 42.6** ±0.6 · #524 of 732 · from 11 benchmark results across 5 sources
Score within the [LMArena (text)](https://whatmodel.sh/?set=lmarena-text&score=set) comparison set only.
## What it's good at
● marks DeepSeek-Coder 33B Base among every model on that task or test, worst to best.
task vs the others rank counts tests
coding ▁▁▁▃●▇█▇▅▄▃▁ #377 of 469 84% 6
math ▁▁▃▅●▆▄█▄▂▁▁ #262 of 431* 3.3% 1
coding · #377 of 469 · 84% of its score
Beats [LFM2.5 8B-A1B](lfm2-5-8b-a1b.md), [Qwen3-VL 2B Thinking](qwen3-vl-2b-thinking.md), [DeepSeek R1 Distill Qwen 7B](deepseek-r1-distill-qwen-7b.md) here, though they rank higher overall. Trails [Qwen2.5 1.5B Instruct](qwen2-5-1-5b.md), [WizardLM 13B V1.2](wizardlm-13b-v1-2.md), [Granite 3.0 2B Instruct](granite-3-0-2b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
HumanEval 51.2% ▁·▁▁▃▄●▅█▃▁▁ beats 36/99 easy · avg 76% 53% sharp · at its level · several runs
HumanEval+ 44.5% ▁·▁▁▁▃▄●▄█▃▁ beats 36/99 easy · avg 70% 21% sharp · at its level
DS-1000 40.2% ▄▄·▄██▄█·█·● beats 12/12 medium · avg 49% 3.6% at its level
MBPP 73.2% ▄▄▄██▄·▄█·▄● beats 12/12 easy · avg 81% 3.6% at its level
CRUXEval-I 46.5% ▁··▁▅▂█●▁··▁ beats 18/22 medium · avg 52% 1.3% at its level
CRUXEval-O 48.6% ▁·▂▃▁█▁●···▁ beats 19/22 medium · avg 51% 1.2% blunt · at its level
math · #262 of 431 · 3.3% of its score
Beats [GPT-4o (2024-11-20)](gpt-4o-2024-11-20.md), [GPT-4 Turbo (2024-04-09)](gpt-4-turbo-2024-04-09.md), [Llama 3.1 405B Instruct](llama-3-1-405b.md) here, though they rank higher overall. Trails [DeepSeek R1 Distill Qwen 1.5B](deepseek-r1-distill-qwen-1-5b.md), [Gemma 7B (base)](gemma-7b-base.md), [CodeGemma 7B (v1.0)](codegemma-7b.md) here, though they rank lower overall.
test result vs the others in its run difficulty counts why
GSM8K 58.7% ▄▄▄·▄█·██·●▄ beats 11/12 easy · avg 75% 3.3% sharp · at its level
other tests (2)
test result vs the others in its run difficulty counts why
EvoEval (overall) 36.8% ▂▂▃▄▄█●▄▅▃▁▃ beats 25/42 medium · avg 52% 9.1% at its level
BigCodeBench (Complete) 46.6% ▂▁▁▄▂▅▇█●▂▂▂ beats 45/58 medium · avg 48% 3.7% at its level
source results share of the score
EvalPlus / EvalPerf published evaluatio… 2 48% ▇▇▇▇▇▇
Top Leaderboard Ranking = Top Coding Pr… 2 25% ▇▇▇
StarCoder 2 and The Stack v2: The Next … 4 21% ▇▇
BigCodeBench paper 1 4% ▇
CRUXEval published evaluation results 2 2% ▇
prior: pull toward the middle 0% ▇
weighted average = 42.6
Licences: EvalPlus / EvalPerf published evaluation runs ([Apache-2.0](../../licenses/evalplus-published-runs-Apache-2.0.txt)) · Top Leaderboard Ranking = Top Coding Proficiency ([CC BY 4.0](../../licenses/paper-2403.19114v1-CC-BY-4.0.txt)) · StarCoder 2 and The Stack v2: The Next Generation ([CC BY-SA 4.0](../../licenses/paper-2402.19173v1-CC-BY-SA-4.0.txt)) · BigCodeBench paper ([CC BY 4.0](http://creativecommons.org/licenses/by/4.0/)) · CRUXEval published evaluation results ([MIT](../../licenses/cruxeval-published-results-MIT.txt))
## How to run it
### llama.cpp (recommended)
1. Install llama.cpp. Homebrew and winget installs name the commands llama-cli and llama-server instead. [Other ways to install](https://github.com/ggml-org/llama.cpp/blob/master/docs/install.md)
```sh
# macOS / Linux
curl -LsSf https://llama.app/install.sh | sh
# Windows
irm https://llama.app/install.ps1 | iex
```
2. Run DeepSeek-Coder 33B Base (Q4_K_M, needs ~19.9 GB, from [huggingface.co/TheBloke](https://huggingface.co/TheBloke/deepseek-coder-33B-base-GGUF/tree/1bb6bd21215f0897fdfd11a0a6b119c42da866b3))
```sh
llama cli -hf TheBloke/deepseek-coder-33B-base-GGUF:Q4_K_M
```
3. Or start a local server and open the address it prints (http://localhost:9931)
```sh
llama serve -hf TheBloke/deepseek-coder-33B-base-GGUF:Q4_K_M
```
Other versions: Q8_0 (~35.4 GB) `llama cli -hf TheBloke/deepseek-coder-33B-base-GGUF:Q8_0`
### LM Studio
1. Install LM Studio. Open the app once before using the lms command. [Download it (macOS / Windows / Linux)](https://lmstudio.ai/download)
2. Run DeepSeek-Coder 33B Base (Q4_K_M, needs ~19.9 GB, from [huggingface.co/TheBloke](https://huggingface.co/TheBloke/deepseek-coder-33B-base-GGUF/tree/1bb6bd21215f0897fdfd11a0a6b119c42da866b3))
```sh
lms get https://huggingface.co/TheBloke/deepseek-coder-33B-base-GGUF@Q4_K_M -y
```
3. Then run lms ls to see its name, and lms chat followed by that name
Other versions: Q8_0 (~35.4 GB) `lms get https://huggingface.co/TheBloke/deepseek-coder-33B-base-GGUF@Q8_0 -y`
### Ollama
1. Install Ollama. [Download it (macOS / Windows)](https://ollama.com/download)
```sh
# Linux
curl -fsSL https://ollama.com/install.sh | sh
```
2. Run DeepSeek-Coder 33B Base (Q4_K_M, needs ~20 GB, from [ollama.com](https://ollama.com/library/deepseek-coder/tags))
```sh
ollama run deepseek-coder:33b-base-q4_K_M
```
Other versions: Q8_0 (~35 GB) `ollama run deepseek-coder:33b-base-q8_0`
Model repository and documentation: [huggingface.co/deepseek-ai](https://huggingface.co/deepseek-ai/deepseek-coder-33b-base)
[interactive version](https://whatmodel.sh/model/deepseek-coder-33b-base) · [JSON](https://whatmodel.sh/api?format=json&all=1)